multiColumnarLearnerExample.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.column_data import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.random.random([100,1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(100, 1)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = np.concatenate([x[1:-2], x[2:-1], x[3:]], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(97, 3)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = x[:-3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "trn_ds = ColumnarDataset.from_data_frames(pd.DataFrame(), pd.DataFrame(x[:-30]),y[:-30], is_multi=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "val_ds = ColumnarDataset.from_data_frames(pd.DataFrame(), pd.DataFrame(x[-30:]),y[-30:], is_multi=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "md = ColumnarModelData('', trn_ds, val_ds,4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[\n",
       "     0\n",
       "     0\n",
       "     0\n",
       "     0\n",
       "     0\n",
       "     0\n",
       "     0\n",
       "     0\n",
       " [torch.FloatTensor of size 8x1], \n",
       "  0.0004\n",
       "  0.3039\n",
       "  0.7304\n",
       "  0.1238\n",
       "  0.5439\n",
       "  0.5861\n",
       "  0.0922\n",
       "  0.4759\n",
       " [torch.FloatTensor of size 8x1], \n",
       " (0 ,.,.) = \n",
       "   0.3039  0.7304  0.1238\n",
       " \n",
       " (1 ,.,.) = \n",
       "   0.7304  0.1238  0.5439\n",
       " \n",
       " (2 ,.,.) = \n",
       "   0.1238  0.5439  0.5861\n",
       " \n",
       " (3 ,.,.) = \n",
       "   0.5439  0.5861  0.0922\n",
       " \n",
       " (4 ,.,.) = \n",
       "   0.5861  0.0922  0.4759\n",
       " \n",
       " (5 ,.,.) = \n",
       "   0.0922  0.4759  0.0461\n",
       " \n",
       " (6 ,.,.) = \n",
       "   0.4759  0.0461  0.5027\n",
       " \n",
       " (7 ,.,.) = \n",
       "   0.0461  0.5027  0.6657\n",
       " [torch.FloatTensor of size 8x1x3]]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "next(iter(md.val_dl))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "m = md.get_learner([],1,0,3,[10],[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "00d6dbdc8eda40458bbd257385536e2a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/html": [
       "<p>Failed to display Jupyter Widget of type <code>HBox</code>.</p>\n",
       "<p>\n",
       "  If you're reading this message in the Jupyter Notebook or JupyterLab Notebook, it may mean\n",
       "  that the widgets JavaScript is still loading. If this message persists, it\n",
       "  likely means that the widgets JavaScript library is either not installed or\n",
       "  not enabled. See the <a href=\"https://ipywidgets.readthedocs.io/en/stable/user_install.html\">Jupyter\n",
       "  Widgets Documentation</a> for setup instructions.\n",
       "</p>\n",
       "<p>\n",
       "  If you're reading this message in another frontend (for example, a static\n",
       "  rendering on GitHub or <a href=\"https://nbviewer.jupyter.org/\">NBViewer</a>),\n",
       "  it may mean that your frontend doesn't currently support widgets.\n",
       "</p>\n"
      ],
      "text/plain": [
       "HBox(children=(IntProgress(value=0, description='Epoch', max=1), HTML(value='')))"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch      trn_loss   val_loss                                                                                         \n",
      "    0      0.226184   0.08783   \n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[array([0.08783])]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "m.fit(1,1)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}